siteforge 0.1.3

Archive websites into AI-readable local knowledge archives
Documentation
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
use std::collections::{BTreeMap, HashMap};
use std::sync::{Arc, Mutex};
use std::time::{Duration, Instant};

use chrono::{DateTime, Utc};
use reqwest::header::{HeaderMap, HeaderName, HeaderValue, CONTENT_TYPE, COOKIE, USER_AGENT};
use robots_txt::{matcher::SimpleMatcher, parts::RequestRate, Robots};
use serde::{Deserialize, Serialize};
use tokio::time::sleep;
use url::Url;

use crate::config::Config;
use crate::errors::{Result, SiteforgeError};

#[derive(Debug, Clone)]
pub struct HttpFetcher {
    client: reqwest::Client,
    user_agent: String,
    retry_count: usize,
    delay: Duration,
    robots_cache: Arc<Mutex<HashMap<String, String>>>,
    last_request_start: Arc<Mutex<HashMap<String, Instant>>>,
    extra_headers: HeaderMap,
}

#[derive(Debug, Clone, Serialize, Deserialize)]
pub struct FetchResult {
    pub requested_url: Url,
    pub final_url: Url,
    pub status: u16,
    pub headers: BTreeMap<String, String>,
    pub mime_type: Option<String>,
    pub fetched_at: DateTime<Utc>,
    pub content_hash: String,
    pub bytes: Vec<u8>,
}

impl HttpFetcher {
    pub fn new(
        config: &Config,
        delay_ms: u64,
        headers: &[(String, String)],
        cookie: Option<&str>,
    ) -> Result<Self> {
        let client = reqwest::Client::builder()
            .user_agent(config.user_agent.clone())
            .timeout(Duration::from_secs(config.timeout_secs))
            .redirect(reqwest::redirect::Policy::limited(10))
            .build()?;
        Ok(Self {
            client,
            user_agent: config.user_agent.clone(),
            retry_count: config.retry_count,
            delay: Duration::from_millis(delay_ms),
            robots_cache: Arc::new(Mutex::new(HashMap::new())),
            last_request_start: Arc::new(Mutex::new(HashMap::new())),
            extra_headers: build_extra_headers(headers, cookie)?,
        })
    }

    pub async fn fetch(&self, url: Url) -> Result<FetchResult> {
        let robots_delay = self.ensure_robots_allowed(&url).await?;
        self.wait_for_origin(&url, robots_delay).await;

        let mut last_error = None;
        for attempt in 0..=self.retry_count {
            match self.fetch_once(url.clone()).await {
                Ok(result) if should_retry_status(result.status) && attempt < self.retry_count => {
                    last_error = Some(format!("HTTP {}", result.status));
                    sleep(backoff(attempt)).await;
                }
                Ok(result) => return Ok(result),
                Err(err) if attempt < self.retry_count => {
                    last_error = Some(err.to_string());
                    sleep(backoff(attempt)).await;
                }
                Err(err) => return Err(err),
            }
        }

        Err(SiteforgeError::message(
            last_error.unwrap_or_else(|| format!("failed to fetch {url}")),
        ))
    }

    pub async fn robots_sitemaps(&self, url: &Url) -> Result<Vec<Url>> {
        let robots_txt = self.robots_text(url).await?;
        Ok(sitemap_urls_from_robots(url, &robots_txt))
    }

    async fn fetch_once(&self, url: Url) -> Result<FetchResult> {
        let response = self.request(url.clone()).send().await?;
        let status = response.status().as_u16();
        let final_url = response.url().clone();
        let headers = headers_to_map(response.headers());
        let mime_type = response
            .headers()
            .get(CONTENT_TYPE)
            .and_then(|value| value.to_str().ok())
            .map(|value| {
                value
                    .split(';')
                    .next()
                    .unwrap_or(value)
                    .trim()
                    .to_ascii_lowercase()
            });
        let bytes = response.bytes().await?.to_vec();
        let content_hash = blake3::hash(&bytes).to_hex().to_string();
        Ok(FetchResult {
            requested_url: url,
            final_url,
            status,
            headers,
            mime_type,
            fetched_at: Utc::now(),
            content_hash,
            bytes,
        })
    }

    async fn ensure_robots_allowed(&self, url: &Url) -> Result<Option<Duration>> {
        let origin = origin_key(url);
        let robots_txt = self.robots_text_for_origin(url, &origin).await?;

        if robots_txt.trim().is_empty() {
            return Ok(None);
        }

        let robots = Robots::from_str_lossy(&robots_txt);
        let section = robots.choose_section(&self.user_agent);
        let matcher = SimpleMatcher::new(&section.rules);
        let robots_delay = [
            duration_from_crawl_delay(section.crawl_delay),
            duration_from_request_rate(section.req_rate),
        ]
        .into_iter()
        .flatten()
        .max();
        let mut path = url.path().to_string();
        if let Some(query) = url.query() {
            path.push('?');
            path.push_str(query);
        }
        if matcher.check_path(&path) {
            Ok(robots_delay)
        } else {
            Err(SiteforgeError::RobotsDenied(url.to_string()))
        }
    }

    async fn fetch_robots(&self, url: &Url) -> Result<String> {
        let robots_url = robots_url(url)?;
        let response = self.request(robots_url).send().await?;
        if !response.status().is_success() {
            return Ok(String::new());
        }
        Ok(response.text().await?)
    }

    async fn robots_text(&self, url: &Url) -> Result<String> {
        let origin = origin_key(url);
        self.robots_text_for_origin(url, &origin).await
    }

    async fn robots_text_for_origin(&self, url: &Url, origin: &str) -> Result<String> {
        let cached = self
            .robots_cache
            .lock()
            .expect("robots cache poisoned")
            .get(origin)
            .cloned();
        match cached {
            Some(value) => Ok(value),
            None => {
                let value = self.fetch_robots(url).await.unwrap_or_default();
                self.robots_cache
                    .lock()
                    .expect("robots cache poisoned")
                    .insert(origin.to_string(), value.clone());
                Ok(value)
            }
        }
    }

    async fn wait_for_origin(&self, url: &Url, robots_delay: Option<Duration>) {
        let min_delay = robots_delay
            .map(|delay| delay.max(self.delay))
            .unwrap_or(self.delay);
        if min_delay.is_zero() {
            return;
        }

        let origin = origin_key(url);
        loop {
            let sleep_for = {
                let mut starts = self
                    .last_request_start
                    .lock()
                    .expect("request pacing state poisoned");
                let now = Instant::now();
                match starts.get(&origin).copied() {
                    Some(last) => {
                        let elapsed = now.saturating_duration_since(last);
                        if elapsed >= min_delay {
                            starts.insert(origin.clone(), now);
                            None
                        } else {
                            Some(min_delay - elapsed)
                        }
                    }
                    None => {
                        starts.insert(origin.clone(), now);
                        None
                    }
                }
            };

            match sleep_for {
                Some(duration) => sleep(duration).await,
                None => break,
            }
        }
    }

    fn request(&self, url: Url) -> reqwest::RequestBuilder {
        let mut request = self
            .client
            .get(url)
            .header(USER_AGENT, self.user_agent.clone());
        for (name, value) in &self.extra_headers {
            request = request.header(name, value);
        }
        request
    }
}

pub fn is_html(result: &FetchResult) -> bool {
    result
        .mime_type
        .as_deref()
        .map(|mime| mime.contains("html") || mime == "text/plain")
        .unwrap_or_else(|| looks_like_html(&result.bytes))
}

pub fn bytes_to_string(bytes: &[u8]) -> String {
    String::from_utf8_lossy(bytes).into_owned()
}

fn looks_like_html(bytes: &[u8]) -> bool {
    let sample = String::from_utf8_lossy(&bytes[..bytes.len().min(512)]).to_ascii_lowercase();
    sample.contains("<html") || sample.contains("<!doctype html") || sample.contains("<article")
}

fn should_retry_status(status: u16) -> bool {
    matches!(status, 408 | 425 | 429 | 500 | 502 | 503 | 504)
}

fn backoff(attempt: usize) -> Duration {
    Duration::from_millis(500 * 2u64.saturating_pow(attempt as u32))
}

fn duration_from_crawl_delay(seconds: Option<f64>) -> Option<Duration> {
    seconds.and_then(|seconds| {
        (seconds.is_finite() && seconds > 0.0).then(|| Duration::from_secs_f64(seconds.min(3600.0)))
    })
}

fn duration_from_request_rate(rate: Option<RequestRate>) -> Option<Duration> {
    rate.and_then(|rate| {
        (rate.requests > 0 && rate.seconds > 0)
            .then(|| Duration::from_secs_f64(rate.seconds as f64 / rate.requests as f64))
    })
}

fn build_extra_headers(headers: &[(String, String)], cookie: Option<&str>) -> Result<HeaderMap> {
    let mut map = HeaderMap::new();
    for (name, value) in headers {
        let name = HeaderName::from_bytes(name.as_bytes()).map_err(|err| {
            SiteforgeError::message(format!("invalid header name {name:?}: {err}"))
        })?;
        let value = HeaderValue::from_str(value).map_err(|err| {
            SiteforgeError::message(format!("invalid value for header {name:?}: {err}"))
        })?;
        map.insert(name, value);
    }
    if let Some(cookie) = cookie.filter(|cookie| !cookie.trim().is_empty()) {
        let value = HeaderValue::from_str(cookie).map_err(|err| {
            SiteforgeError::message(format!("invalid cookie header value: {err}"))
        })?;
        map.insert(COOKIE, value);
    }
    Ok(map)
}

fn headers_to_map(headers: &HeaderMap) -> BTreeMap<String, String> {
    headers
        .iter()
        .filter_map(|(key, value)| {
            value
                .to_str()
                .ok()
                .map(|value| (key.as_str().to_ascii_lowercase(), value.to_string()))
        })
        .collect()
}

fn origin_key(url: &Url) -> String {
    match url.port() {
        Some(port) => format!(
            "{}://{}:{port}",
            url.scheme(),
            url.host_str().unwrap_or_default()
        ),
        None => format!("{}://{}", url.scheme(), url.host_str().unwrap_or_default()),
    }
}

fn robots_url(url: &Url) -> Result<Url> {
    let mut robots = url.clone();
    robots.set_path("/robots.txt");
    robots.set_query(None);
    robots.set_fragment(None);
    Ok(robots)
}

fn sitemap_urls_from_robots(base: &Url, robots_txt: &str) -> Vec<Url> {
    let mut urls = robots_txt
        .lines()
        .filter_map(|line| line.split('#').next())
        .filter_map(|line| line.split_once(':'))
        .filter(|(key, _)| key.trim().eq_ignore_ascii_case("sitemap"))
        .filter_map(|(_, value)| {
            Url::parse(value.trim())
                .or_else(|_| base.join(value.trim()))
                .ok()
        })
        .collect::<Vec<_>>();
    urls.sort();
    urls.dedup();
    urls
}

#[cfg(test)]
mod tests {
    use super::*;

    #[test]
    fn parses_positive_crawl_delay() {
        assert_eq!(
            duration_from_crawl_delay(Some(1.5)),
            Some(Duration::from_millis(1500))
        );
        assert_eq!(duration_from_crawl_delay(Some(0.0)), None);
        assert_eq!(duration_from_crawl_delay(Some(f64::NAN)), None);
    }

    #[test]
    fn parses_request_rate_delay() {
        assert_eq!(
            duration_from_request_rate(Some(RequestRate::new(2, 5))),
            Some(Duration::from_millis(2500))
        );
        assert_eq!(
            duration_from_request_rate(Some(RequestRate::new(0, 5))),
            None
        );
    }

    #[test]
    fn origin_key_preserves_non_default_port() {
        let url = Url::parse("https://example.test:8443/path").unwrap();
        assert_eq!(origin_key(&url), "https://example.test:8443");
    }

    #[test]
    fn builds_authorized_request_headers() {
        let headers = build_extra_headers(
            &[("Authorization".to_string(), "Bearer token".to_string())],
            Some("session=abc"),
        )
        .unwrap();
        assert_eq!(headers.get("authorization").unwrap(), "Bearer token");
        assert_eq!(headers.get(COOKIE).unwrap(), "session=abc");
    }

    #[test]
    fn rejects_invalid_request_headers() {
        let err = build_extra_headers(&[("Bad Header".to_string(), "value".to_string())], None)
            .unwrap_err()
            .to_string();
        assert!(err.contains("invalid header name"));
    }

    #[test]
    fn extracts_sitemaps_from_robots() {
        let base = Url::parse("https://example.test/docs/").unwrap();
        let urls = sitemap_urls_from_robots(
            &base,
            "User-agent: *\nSitemap: /sitemap.xml\nSitemap: https://cdn.example.test/sitemap.xml\n",
        );
        assert_eq!(urls.len(), 2);
        assert!(urls
            .iter()
            .any(|url| url.as_str() == "https://example.test/sitemap.xml"));
    }
}